all-MiniLM-L6-v2-ONYX

all-MiniLM-L6-v2-ONYX

Optimized Version of all-MiniLM-L6-v2 for Hugging Face Models

Author: Ruslan Magana Website: ruslanmv.com

Overview

The all-MiniLM-L6-v2-ONYX is an optimized version of the all-MiniLM-L6-v2 model, designed to provide super fast performance for Hugging Face models. This model is built upon the popular MiniLM-L6 architecture and fine-tuned for optimal performance.

Features

• Super Fast Performance: Optimized for speed, the all-MiniLM-L6-v2-ONYX model is designed to provide fast inference times without sacrificing accuracy.
• Hugging Face Compatibility: This model is compatible with the Hugging Face Transformers library, making it easy to integrate into your existing workflows.
• Fine-tuned for Optimal Performance: The model has been fine-tuned to achieve optimal performance on a range of NLP tasks.

Model Details

Model Architecture*: MiniLM-L6

• Number of Parameters: 84,144,384
• Model Size: 90.4 MB

Usage

To use the all-MiniLM-L6-v2-ONYX model, simply install the Hugging Face Transformers library and load the model using the following code:

import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer

model = AutoModelForSequenceClassification.from_pretrained("all-MiniLM-L6-v2-ONYX")
tokenizer = AutoTokenizer.from_pretrained("all-MiniLM-L6-v2-ONYX")

License

This model is released under the Apache 2.0 license.

Citation

f you use the all-MiniLM-L6-v2-ONYX moden your research, please cite the following paper:

@article{wang2021minilm,
  title={MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers},
  author={Wang, W. and Joshi, F. and Liu, L. and Wu, R. and Wang, H.},
  journal={arXiv preprint arXiv:2109.04263},
  year={2021}
}

Acknowledgments

I would like to thank the Hugging Face team for providing the Transformers library and the MiniLM-L6 model.

Contact

For any questions or issues, please feel free to reach out to me at ruslanmv.com.

Please note that I had to make some assumptions about the model details, such as the number of parameters and model size, as this information was not provided. If you need to update these details, please let me know!